In order to quickly identify loads of vehicle on a bridge, the effects of moving loads on the bridge strain and deflection are analyzed based on the Euler beam model. The artificial neural network (ANN) is used in bridge moving loads identification. The moving loads are identified on a 30-width simply-supported bridge. The influences of different activation function combinations and algorithm on the network are discussed. The identification results of different load conditions are analyzed and the effect of noise is considered. Finally, the rationality of the method is verified by experiments. The results show that transfer function has little influence on the recognition result, but training algorithm has a great influence on the recognition result. By selecting the appropriate network structure and training methods, the BP neural network can identify the location, speed and dynamic loads of the vehicle according to the strain response of the bridge.